Experts have massively underestimated solar. Why? (Solar: Part 1)

This is part 1 of a three part series.  Look here for Part 2 and Part 3.

Last year the joists undergirding our carbon-fueled energy system shuddered and flexed.  But in the US, barely anyone noticed.

From Forbes:

The latest battering to coal’s standing came when Dubai announced June 27 that it would build a massive 800-megawatt solar plant that will produce electricity at an average cost of 2.99 cents a kilowatt hour, substantially below what even coal-fired power plants charge.

The price of 2.99 ¢/kWh was 30% cheaper than coal.  It was half the price of a solar bid in Dubai just 18 months earlier.  It was without subsidies.

And it’s not a fluke.

Experts have declared that solar will not unseat fossil fuels for decades, if ever.  The experts are wrong.  Solar has transformed from folk song to anthem, and within the next decade will change the way we talk about electricity, industry, and global politics.

But that’s getting ahead of ourselves.

Solar’s rise has been swift and nearly unbroken, but because the global energy system is so vast most of us have barely noticed.  Today’s solar surge is a story about the power of technology, and the power of continuous and relentless engineering.

But most of all, the story of solar is about belief – a belief in what can be done.  In solar, no one – not even environmentalists – believed it would grow at such a torrid pace, save those who were working in the industry themselves.  But that was enough.

This blog post is in three parts: First, I recount what the experts got wrong, using numbers from utility grade solar (rooftop solar has a lot more variables involved).  Many of the goals for solar that you think impossible are in fact happening now — just not in the US.

Secondly, I decipher why we collectively blew this prediction.  Understanding the growth of technology is hard: it’s part engineering, part economics, and a large heaping of pure grit.  When the people on the ground believe in their future, but the forecasters do not, the forecasters will be surprised.

Third, I talk about the impacts of a world where solar is cheaper than coal, starting today.  This last bit involves some speculation, which makes me nervous.  But if I am asking you to throw away your framework for understanding the present, we must stand a new structure in its place.

You have heard that solar can’t have an impact on climate change, that the Chinese are dragging their feet on cleaning up their emissions, and that fossil fuels are going to remain our main source of energy for the future.  Let’s hear a new story.

And like all good stories, this one starts in LA.

ARCO’s 1 MW solar plant in Carrizo Plains, CA.  Photo: Tribune


Solar is growing much, much faster than you think

In the 1980s, a brown haze shrouded the sky above Interstate 10 on the drive eastward out of Los Angeles .  Called the ‘Valley of Smoke’ by Native Americans, LA is bounded by the San Gabriel mountains to the north and the San Bernardino mountains east; as the winds blow inland over the hills, soot and ash in the air remains trapped in the valley, with no route for escape.  But further north on I-15, on the other side of the Cajon pass, beckon the searing, cloudless skies of the Mojave Desert.

The city immediately north of Cajon, Hesperia, California, was thus the ideal site for LA-based ARCO to build the world’s first 1 MW (megawatt) solar plant in 1982.  Small by energy standards, it could power just 800 homes.  Yet it was seen as the just first step in a long revolution in energy, which would bring clean, inexpensive power to the masses.

ARCO, an oil company, was spooked by climbing oil prices in the 1970s.  If oil were to rise from $40 to $60 a barrel ($140 a barrel in today’s money), and if government tax credits were extended, solar could be big business.   After mastering construction of its Hesperia installation, ARCO built a second plant in 1983 in Carrizo Plain, CA, an even sunnier location; at 5.2 MW, it was by far the largest in the world.

But oil prices fell. The Reagan administration’s interests turned away from solar, and tax credits vanished.  In 1990, ARCO auctioned its solar plants to investors, who disassembled the panels, selling them for scrap.  Solar was valuable in places like the Caribbean, islands where fossil fuels were scarce.  In California, solar was a toy for hippies, and little else.

This represents, more or less, the world’s expectations for solar ever since.

A few months ago, things changed.  In August, a solar plant in Chile bid at 2.91¢ per kWh, several cents cheaper that Dubai’s record price.  In September a 350 MW plant in Abu Dhabi received a bid at 2.42¢ per kWh.  Abu Dhabi ended up agreeing to a slightly higher price – as part of a deal to expand the plant by 3-fold, to 1.1 GW (a GW is 1,000 MW).

The world is starting to believe in solar.

A new solar plant is suddenly not just cheaper than the cost of a new coal plant.  The cost of financing a solar plant, in sunny climates, now competes with the costs of mining and transporting coal.  In these conditions (and note that they vary by geography), the economically sensible thing to do is to mothball operating coal plants, and replace them with solar.

Record low bids.png
Record low bids for solar.  Bid data: Cleantechies.  Natural gas prices and coal prices, conversion to kWh and transport prices for different sources of coal all using data from US Energy Information Administration.


This scenario was fantastical just a few years ago.  In 2014, Google engineers – formerly of their discarded renewable energy team – said this about the dim prospects for wind and solar:

Consider an average U.S. coal or natural gas plant that has been in service for decades; its cost of electricity generation is about 4 to 6 U.S. cents per kilowatt-hour. Now imagine what it would take for the utility company that owns that plant to decide to shutter it and build a replacement plant using a zero-carbon energy source. The owner would have to factor in the capital investment for construction and continued costs of operation and maintenance — and still make a profit while generating electricity for less than $0.04/kWh to $0.06/kWh.

The unthinkable has happened.  The hippies have won.  Renewables are cheaper than coal in Chile and the Middle East, and will soon become cheaper in other regions as well.  The world has a good chance of meeting the Paris Climate Agreement’s initial goal of limiting warming to a 2°C rise.

If that surprises you, it’s probably because you have been listening to experts.  Experts who said there is no way solar could scale fast enough to make a meaningful impact on energy generation or global warming before 2030, or 2050.

Here’s how the experts got it wrong.

The art of misforecasting

The International Energy Agency has always been a conservative lot.  Their stable of experts came of age in a world powered by oil and gas, a world where politics had a far greater influence on the price of energy than technology.  Still, it is astonishing to see just how poor a job of forecasting they have done.  (Note the star is my own estimate for 2020; I’ll explain the method I used to arrive at it below.)

Original graph: Bloomberg.  Dotted line and star added by me.  WEO stands for “World Energy Outlook”, a publication from the IEA.


But let’s be fair – it wasn’t just the IEA who has underestimated solar.

Since 2007, the environmental organization Greenpeace has published several manifestos for our energy future.  In them, Greenpeace lays out an aggressive timeline for investment in all renewables, asking the world to struggle to the edge of what they believe is achievable.  And they, too, have had to revise up their estimates for solar with each new publication.


Data from Greenpeace Energy [R]evolution reports from 2007, 2010, and 2015.


In the US, renewable energy is barely discussed amidst a larger partisan war, where reality is secondary to political considerations.  But this summer’s price drop in solar got the attention of China, the world’s largest energy consumer:

“China will invest 2.5 trillion yuan ($360 billion) into renewable power projects by 2020 as the world’s largest consumer of energy starts to move away from coal…  The five-year plan follows last month’s announcement… that 1 trillion yuan ($145 billion) would be invested to increase China’s solar capacity five-fold by 2020”  

Newsweek, January 5, 2017

Let’s put these numbers in perspective.  The world has about 280 GW of solar installed in total; of this, about 70GW is in China.  This quote implies that China alone will install another 280 GW in the next four years.

If the estimate of $145 billion in investment is accurate (and since most of the world’s solar production is in China, you can bet that the number was scrubbed with suppliers before-hand), that 280 GW would produce power at about $0.50 per W installed, or less than 2¢ per kWh in sunny climates.  That makes solar as cheap as coal in China, a country with few of the West’s environmental restrictions.

And China is not alone.  Here is how India reacted to the falling prices:

“The Indian government has forecast that it will exceed the renewable energy targets set in Paris last year by nearly half[,] and three years ahead of schedule.  A draft 10-year energy blueprint published this week predicts that 57% of India’s total electricity capacity will come from non-fossil fuel sources by 2027. The Paris climate accord target was 40% by 2030”

The Guardian, Dec 22, 2016

If India is to be believed, this would amount to $40B of further investment in solar.  And this is credible – they already have a $20B commitment from Japan’s Softbank, Taiwan’s Foxconn, and India’s Bharti Enterprises.  It’s credible because solar is becoming as cheap as coal.

The US led the world in solar investment in the 1970s, and the 1980s, and the 1990s.  But more recently, we have objected to investing in renewable energy to save the planet, in part because we believed that with China and India polluting unchecked, the planet could not be saved.  Solar was a toy for the rich; developing nations had to burn coal to catch up to us economically.  And until now, they have.

Two things have changed.  First, smog from coal has choked cities like Delhi (the most polluted city in the world) and Beijing, and city dwellers are revolting.  Global climate change is abstract; a generation of children with chronic asthma provides focus.

Second, solar has, just now, become economical.  Below is a chart of  the price of solar modules, which have historically made up the majority of cost in a solar installation.  The vertical axis is price; the horizontal axis is the total amount of solar panels that have been shipped in the history of the world.  Note this plot is on a logarithmic scale – each major grid represents a factor of 10 – and the green line represents what most experts expected for solar.

Swanson's Law.png
Solar module cost plotted against cumulative shipments on a log/log graph.  The relationship is often referred to as Swanson’s Law – I’ll talk about that more in the next section.  Image: Wikipedia.


For the last decade, solar prices have dropped at a far faster than historical rates.  This came as a surprise.  And up until recently, not even Greenpeace believed the trend would continue.

But now Greenpeace believes.  Their latest ambition is for the world to have 850 GW of solar by 2020.   If the world outside China commits to keeping up with China’s investments – and the market is evenly split between China and the rest of the world today – we will in fact end up exactly where Greenpeace aspires.  Heck, just extrapolating the recent trends would lead to a prediction 850 GW by 2020, if not more.

Even with all of this solar installation, China’s grid will only be 15% renewable by 2020.  Beyond 2020, there will be plenty of demand for more.

Yet in the US, we’re oblivious to this trend, in part because our solar economics are different. Installation of modules here costs 3X what it does in China, largely because we have not invested in efficient techniques for installation.  Below is a cost breakdown for solar installation, by country, in 2015.  (Prices have fallen since then, of course).

Balance of system costs (everything but the solar module), by country, for utility grade solar.  Source: International Renewable Energy Agency (IRENA).


It would be comforting to think that our reason for lagging in cost is because the Chinese are giving away land for free, or because they barely pay their labor.  But look again – Germany has costs roughly on par with China.  And nearly every country in Europe is ahead of us.

We lag because policy has not caught up with the cost reality.  We lag because we do not believe.

The irony here is that our global competitiveness depends on access to cheap energy, yet it was those worried about competitiveness who led us to this disadvantage.  We failed to recognize how fast solar would progress.  We failed to see the possibility that economic policies intended to save money would have – at this moment – exactly the opposite effect.

And if we are to make good policy about our solar future, we first should understand how we have been blindsided by our solar present.

What have experts been doing when they’ve made these predictions about solar?  And why have all of them been repeatedly, consistently wrong?

<Part II – How solar progress progresses>

2 thoughts on “Experts have massively underestimated solar. Why? (Solar: Part 1)

Leave a Reply